Hippocampal network oscillations: dissecting the septo-hippocampal circuitry

How the brain encodes information. Our brain processes information in a network of nerve cells. The communication between these nerve cells occurs via specialized structures called synapses. When an electrical signal (action potential) travels along the nerve cell and reaches the synapse then a messenger substance is released from the synapse. This messenger substance is received by a receptor protein in the membrane of the adjacent nerve cell. The firing of action potentials of individual nerve cells and the communication between these cells via synapses is very important for the processing of information, but it might not be sufficient. To allow efficient information processing by individual nerve cells, their action potential firing might have to be embedded in a temporal context. Such a 'clock signal' in the brain might be provided by so called network oscillations. These 'brain waves' arise from voltage changes in the extracellular space of the brain, which are caused by the ion flux that is associated with synchronous activity of large groups of nerve cells. Using electrodes that are inserted into the brain these waves can be measured. It was found that there are different types of brain waves and that each of them is associated with a particular behaviour in an animal. The hippocampus is a brain structure which is important for learning and memory. Here different brain waves are hypothesised to be necessary for the encoding of information in different phases of memory formation. However it has not been possible to directly test this. In the experiments I propose in this application I want to test the function of brain waves in memory formation. We have developed a new technique that allows us to interfere with selected synapses and thus with communication between selected nerve cells that are important for the generation of different types of brain waves. By interfering with nerve cell communication in different parts of the nerve cell network we want to find out which parts of the network generate which type of brain waves. Once we have this information we will selectively disrupt distinct types of brain waves and test how this influences learning and memory.

Synchronous network oscillations might serve as a general mechanism of how the brain encodes information. These oscillations of the extracellular field potential arise from the synchronous activity of large neuronal groups and are believed to provide a temporal reference signal for the processing and storage of information. In the hippocampus distinct frequencies of network oscillations can be recorded with different types of behaviour. However, a causal relationship between different oscillatory patterns and behaviour is difficult to test. Knowledge of the underlying neuronal mechanisms would allow selective perturbations of network oscillations to test their behavioural relevance. We have previously selectively disconnected parvalbumin-positive GABAergic neurons from the fast inhibitory network throughout the entire brain. This manipulation resulted in: 1, reduction of theta oscillations; 2, disruption of cross-frequency coupling between theta and gamma oscillations and 3, disruption of ripple oscillations. To find out where in the distributed network synaptic inhibition of parvalbumin-positive neurons is required for theta and ripple oscillations and theta/gamma coupling we want to selectively perturb synaptic inhibition onto these cells independently in different sub-circuits of the septo-hippocampal network using novel viral and pharmaco-genetic tools with very high spatial and temporal resolution. The effects of these micro-manipulations will be analysed at the cellular and network level and correlated with changes in hippocampal memory performance. With a parallel approach we want to functionally remove parvalbumin-positive neurons from the medial septum to test their role in the generation of theta oscillations and hippocampus-dependent learning.